Benign Familial Neonatal Seizures and Benign Idiopathic Neonatal Seizures

Perrine Plouin

Introduction

In 1989, the International Classification of Epilepsies, Epileptic Syndromes and Related Disorders¹⁵ included two neonatal syndromes among idiopathic conditions in the section on generalized epilepsies and syndromes: benign familial neonatal convulsions and benign neonatal convulsions. At that time, no video-electroencephalogram (EEG) recording of seizures had been reported, and newborn infants were considered as presenting more generalized than focal seizures. Nevertheless, since 2001, it has been proposed to consider at least benign familial neonatal convulsions as a focal clonic or adversive epilepsy. Plouin⁴⁷ proposed that the appellation benign neonatal convulsions was not sufficiently precise: Some symptomatic cases such as hypocalcemia may also be benign, and it would be more accurate if the word idiopathic was added: benign idiopathic neonatal convulsions. Since then, the word “convulsions” has been replaced by the word “seizures” because “convulsions” refer only to the motor components of seizures, excluding the autonomic components. These two syndromes are now called benign familial neonatal seizures (BFNS) and benign idiopathic neonatal seizures (BINS). Only the first remains in the proposed classification because it seems that no new cases of BINS have been reported in the last 15 years.

These two syndromes do not strictly fulfill the criteria for idiopathic generalized epilepsies; the typical trait of generalized spike-and-wave discharge is not present. It can be argued that the immaturity of the central nervous system is responsible for this absence. Nevertheless, since 2001, it has been proposed that at least BFNS be considered a focal clonic or adversive epilepsy.

Benign neonatal seizures are defined by a favorable outcome, that is, normal psychomotor development and the absence of secondary epilepsy. Benign familial neonatal seizures and benign idiopathic neonatal seizures fulfill these criteria, even if some questions remain open.

Recognition of these syndromes allows the prediction of a favorable outcome from the neonatal period, and long-term antiepileptic treatment is not indicated.

Benign Familial Neonatal Seizures

Historical Perspectives

In 1964, Rett and Teubel⁵¹ reported the first BFNS family, with eight cases over three generations. On the third day of life, the male proband developed an initial tonic phase with cyanosis followed by clonic movements of the whole body including the face and eye muscles, and he had 15 to 20 seizure events on the following day. A brother born 16 months later had a similar experience. Several interictal EEGs were reported for these two boys and single EEGs for three other affected relatives. No ictal EEG was recorded. The authors noted the familial history, the normality of the interictal EEG, and the favorable outcome.

In a second family, 14 members in five generations had similar clinical histories.⁹ After a normal delivery, seizures (sometimes with cyanosis) usually started on the third day of life but stopped within 1 month. A few seizures were observed up to 7 months in three children and up to 10 years in two others.

In 1979, Quattlebaum⁵⁰ reported a family in which 11 individuals had seizures that started on or before 3 days of age, 1 at 3 weeks, and 3 at 3 months. Most had seizures until 6 or 8 months, but all were otherwise normal.

Between 1964 and 1989, 26 families were reported, with all authors agreeing on a probable autosomal-dominant inheritance.

In 1989, Leppert et al.³² studied 48 individuals in four generations of a family (Quattelbaum’s family), 19 of them having the characteristics of BFNS. They localized the gene on the long arm of chromosome 20, possibly to 20q13.2, the first linkage to be reported for an epilepsy syndrome. This was soon confirmed in a Newfoundland, Canada, family with 69 affected inviduals,⁵³ in a northern European family,⁵⁵ and in six French families.³⁸ The BFNS syndrome that maps to chromosome 20q has been designated EBN1. Since then, >25 families have been genetically studied in different countries. For 20 of them, the same genetic localization on the long arm of chromosome 20 has been proved.

In 1991, however, Ryan et al.⁵⁵ studied two families with BFNS that did not link to chromosome 20 and concluded that the syndrome of BFNS was genetically heterogeneous. Further study of that family³⁴ demonstrated tight linkage to a locus on 8q, thus verifying heterogeneity (the BFNS syndrome on 8q is designated EBN2). In other families, no linkage could be found on either 20q or 8q.³⁵

In 1995, Anderson et al.¹ characterized a candidate gene for BFNS: the human alpha 4 subunit of the nicotine acetylcholine receptor (hCHRNA4). In 1998, the sequencing of genes that are mutated in EBN1 and EBN2 was reported.⁵⁸ The new gene was named KCNQ2, and analysis of five other ENB1 families yielded two transmembrane missense mutations, two frameshifts, and one splice-site mutation.

A search of the human genome databases for sequences homologous to KCNQ2 detected one in the chromosome 8q24 region.¹⁴ The new gene was named KCNQ3, and a mutant DNA fragment from it was found to segregate perfectly with the EBN2 phenotype in a Mexican-American family.⁵⁸

New EBN1 kindreds have been reported with variations in seizure history or in types of mutation. A wide range of clinical manifestations has been observed, including partial seizures in later life with corresponding focal neurologic deficits. Thus far, however, there is little evidence for a correlation between type of mutation and seizure history.⁸^,³³

The condition of benign familial neonatal seizures is a rare, dominantly inherited epileptic syndrome with a penetrance as high as 85%.

For more details about the genetics of this syndrome see Chapter 18.

Definitions

The diagnosis of BFNS can be considered as a diagnosis by exclusion if every other etiology has been excluded and if there is a family history of neonatal seizures without any severe epilepsy in family members.

Epidemiology

No epidemiologic study of infant or childhood epilepsy has determined the prevalence of BFNS. Among a population of 76 newborns with seizures referred to the neuropediatric unit of Saint Vincent de Paul Hospital between 1994 and 2002 and with video-EEG–recorded seizures, we had only 2 cases of BFNS. The best estimate thus far of the population rate for BFNS comes from a recent prospective, population-based study that involved all obstetric and neonatal units across the province of Newfoundland.⁵² Five cases of BFNS (none of whom were part of the kinship reported by Ronen et al.⁵³) were observed among 34,615 live births from 1 January 1990 to 31 December 1994. Thus the incidence of BFNS was reported as 14.4 per 100,000 live births.

All reported families represent isolated cases. It is probable that families with BFNS are more numerous than would appear from the medical literature; families with this syndrome often know that the outcome is favorable and do not always report the occurrence of seizures to their physician.

Three hundred and fifty-five cases have been reported in the medical literature since the first article by Rett and Teubel.⁵¹ These neonates belong to 44 families. The number of affected generations varies from one to five (Table 1).

Etiology and Basic Mechanisms

By definition, an idiopathic syndrome does not have a specific underlying etiology. This is the case for BFNS.

Basic mechanisms in this syndrome are probably close to those involved in other types of neonatal convulsions. Immature brain is more likely to respond to any kind of injury with epileptic seizures. In this syndrome, genetic susceptibility during the first week of life in full-term neonates is responsible for the appearance of seizures; the cause is a deficit of a specific channelopathy, but the precise mechanism is unknown. The fact that seizures occur in premature infants when they reach 39 to 41 weeks of gestational age means that a step in maturation has to be reached for the channelopathy to be expressed.

Table 1 Benign familial neonatal seizures

Study	Cases per family	Generations per family
Rett and Teubel 1964⁵¹	8	3
Bjerre and Corelius 1968⁹	14	5
Steejohnsen 1968	8	?
Rose and Lombroso 1970⁵⁴	3	3
Goutières 1977²⁶	8	3
Goutières 1977²⁶	2	1
Carton 1978¹³	9	4
Quattelbaum 1979⁵⁰	15^a	4
Pettit and Fenichel 1980⁴⁵	5	3
Tibbles 1980⁵⁹	2	2
	6	2
	3	2
Pavone et al. 1982⁴⁴	7	2
Giacoia 1982²³	7	4
Plouin 1984⁴⁶	2	2
Plouin 1984⁴⁶	2	1
Kaplan and Lacey 1983³¹	12	4
Palencia and Berjon 1985⁴³	4	2
Shevell et al. 1986⁵⁷	5	3
Shevell et al. 1986⁵⁷	4	3
Nieto Barrera et al. 1986⁴¹	14	3
Calero Garcia et al. 1988¹¹	17	4
Cunniff et al. 1988¹⁷	8	2
Giroud et al. 1989²⁴	2	2
	2	2
	2	1
Malafosse et al. 1990³⁷	6^a	3
	4^a	3
	4^a	2
Webb and Bobele 1990⁶³	9	5
Camfield et al. 1991¹²	5	2
Ryan et al. 1991⁵⁵	14^a	3
Ryan et al. 1991⁵⁵	15^b	3
Schiffmann et al. 1991⁵⁶	4	1
Wakai et al. 1991⁶¹	5	2
Aso and Watanabe 1992⁴	3	2
Malafosse et al. 1992³⁸	9^a	4
	3^a	2
	9^a	4
Mami et al. 1993³⁹	4	2
Ronen et al. 1993⁵³	69^a	5
Hirsch et al. 1993³⁰	6^a	2
Bye 1994¹⁰	5	3
Berkovic et al. 1994⁶	10^a	4
Total 1964–1995: 355 cases, 44 families
^aFamilies mapped to chromosome 20 (linkage analysis of Quattelbaum’s family by Leppert et al.³²). ^bFamily mapped to chromosome 8 (linkage analysis of Lewis et al.³⁴).